Velocity-nulling (VN) gradient in GE-EPI.

(a) The diagram of pulse sequence. The VN gradient is highlighted in light blue. (b) The simulated signal attenuation against b values.

Empirical results of the button-press task showing varying levels of draining-vein suppression in different participants.

(a) The paradigm of button-pressing task. (b) The acquired image with 0.9-mm isotropic resolution in three orthogonal views. The 20 layers of M1 are color-coded as shown in the right panel.

(a) BOLD activation maps in the visual cortex without and with NORDIC denoising. (b) BOLD activation maps in the motor cortex without and with NORDIC denoising. The statistical maps were corrected (uncorrected p < 0.05; corrected p < 0.05) and color-coded as indicated by the color bar. (c) Depth-dependent BOLD activation profiles for the M1 regions shown in (b). The shaded areas in light red and blue represent the 95% confidence intervals.

Empirical results of the button-press task showing varying levels of draining-vein suppression in different participants.

(a-d) Depth-dependent BOLD activation profiles in M1. The green contours represent the edges of M1 regions. Each row represents data from a different individual. Columns from left to right correspond to TE = 30 ms, TE = 38 ms, TE = 38 ms with b = 6 s/mm2, and TE = 39 ms with b = 8 s/mm2. The statistical maps were corrected (uncorrected p < 0.05; corrected p < 0.05) and color-coded as indicated by the color bar. The shaded areas in light red and blue represent the 95% confidence intervals (CI).

The activation maps correspond to visual cues from the button-pressing task.

The difference maps were generated by subtracting the activation maps in the middle row (with phase regression) from those in the top row (without phase regression). The blue circles highlight the effective suppression of macrovascular signals by VN gradients.

Impact of VN Gradients on Intra-Cortical FC Matrices.

(a) The panels from left to right display the intra-cortical FC matrix derived from VN fMRI, the intra-cortical FC matrix derived from regular fMRI, and the difference between the two matrices. All FC matrices are Fisher’s z-transformed. (b) Statistical significance of the differences in the FC matrices. Red regions indicate the differences are statistically significant (FDR-corrected p-value < 0.05).

Layer-dependent FC analysis using layer-specific M1 seeds.

(a) Workflow for conducting layer-specific FC analysis. Surface-based labels were converted from individual surface to volume space. Red and blue colors represent the superficial and deep layers, respectively. Seed-based FC analysis was performed in EPI space, followed by morphing to individual surfaces and then to the template surface. (b) Contrast of FC maps corresponding to superficial, middle and deep layers. Warm colors represent stronger FC associated with superficial layers in M1, while cool colors indicate stronger FC associated with deep layers in M1. (c) Differential FC maps showing the magnitude of differences between layers, including superficial vs. middle, deep vs. middle, and superficial vs. deep layers. Abbreviations: Sup – superficial; Mid – middle; S. – superficial; M. – middle; D. – deep.

Layer-dependent FC analysis using layer-specific S1 seeds.

(a) The illustration of surface-based and volume-based S1 labels. (b) Contrast of FC maps corresponding to superficial, middle and deep layers. Warm colors represent stronger FC associated with superficial layers in S1, while cool colors indicate stronger FC associated with deep layers in S1. (c) Differential FC maps showing the magnitude of differences between layers, including superficial vs. middle, deep vs. middle, and superficial vs. deep layers. Abbreviations: Sup – superficial; Mid – middle; S. – superficial; M. – middle; D. – deep.

Layer-specific functional connectivity across functional networks (N = 14).

(a) Fisher’s z transformed FC matrices. The FC matrix on the left represents the Fisher’s z values without a VN gradient, while the matrix on the right corresponds to the Fisher’s z values with a VN gradient. From (b) to (d) displays the depth-dependent FC matrices for a representative ROI pair in visual, sensorimotor and default-mode networks respectively. T-values are color-coded as indicated by the color bar. The left panel displays the matrix without a VN gradient, and the right panel presents the matrix with a VN gradient. Abbreviations: V1 – primary visual cortex; V2 – secondary visual cortex; S1 – primary sensory cortex; vPCC – ventral posterior cingulate cortex; AG – angular gyrus; Vis – visual network; SM – sensorimotor network; dAtt – dorsal attention network; vAtt – ventral attention network; Lim – limbic network; FP – frontoparietal network; DMN – default-mode network.

List of brain-wide layer-dependent fMRI methods.

The imaging protocols need to cover more than half of the human brain to be classified as brain-wide acquisitions.

Generation of layer-specific functional connectivity.

(a) The surface-based Shen268 functional parcellation. (b) The depth-dependent functional connectivity matrix. Abbreviations: Vis – visual network; SM – sensorimotor network; dAtt – dorsal attention network; vAtt – ventral attention network; Lim – limbic network; FP – frontoparietal network; DMN – default-mode network.